Model Reference Adaptive Control of Industrial Robots with Actuator Dynamics

摘要

Abstract An industrial robot is composed of a multidegree-of-freedom mechanism, associated actuator subsystems, and a computer based control system. In general, it is difficult to accurately control the motion of mechanical manipulators by the usual linear control methods, because of the complex nonlinearities of mechanical dynamics and parameter uncertainty. Strong coupling among the actuator subsystems occurs, due to the influence of the inertial forces in the mechanical parts which causes interaction between links. Thus, when the nonlinearities are essential, linear approximations give poor results and the complete dynamic models have to be considered. In the present work the complete nonlinear dynamic model of the robotic manipulator is considered together with its joint actuator (i.e., d.c. motor) dynamics. A control law of the robotic system is developed based on the well known “model reference adaptive control” (MRAC) approach. An exhaustive study of a decoupled reference model of the whole robotic system is proposed which possesses strong stability properties. The paper is completed by some simulation results on trajectory following control of a 3-link manipulator.